Multi-modal multi-resolution atlas of the human neonatal cerebral cortex based on microstructural similarity

In this work, we aimed to generate a comprehensive parcellation of the human neonatal cortex based on multi-modal MRI features. We collected the dataset from the developing human connectome project and estimated ten different MRI features to calculate the similarity between different locations in the neonatal cortex. We developed an automated algorithm based on gradient of the integrated similarity map to generate parcellations at different resolutions. We also provided a manual parcellation based on the multimodal similarity for higher anatomical interpretability. The present work may facilitate structural-functional connectome analysis in early brain development.